As a member of the Dutch Open Language Initiative 'OpenTaal'
(www.opentaal.org), linked to the dutch ooo localisation group, I was
directed to this new initiative.
And a nice one it is. It would contribute a lot in harvesting words.
Which is my contribution to the Dutch Open language Initiative.
For now, we are harvesting words form all kinds of well-edited
document-like archives, like newspapers. This new tool is promising to
contribute quite a bit.
My thoughts about the design one by one:
1) All i saw so far look very good !
2) Besides collection of the words, to have the context it was added
from would be nice (for judgement)
3) a one-time consent ot the user, and then automatic send at avery
addition, is that feasible ? (easier for the user) A one time entering
of the users specialism (like law) would be enough.
4) Is the design such that all data has to be collected on one server ?
That could become huge ... And it requires further distribution of the
collected data.
Or is the design intended in such a way that every language/locale has a
server to send the data to ? I personally think the latter is the better
... But it requires some coordination and a config file for servers/locales.
5) Upon collection the data on the server, it might be usefull to keep a
track record of the words collected, using data like :
- every user(or ip)-word combination stored separately, with date
entered (multiple times foudn words are likely more usefull then others.
date is to keep track of fashion words...
My biggest question is #4, the overall structure of the flow of the
words from the user to the localisation projects. What are the
thoughts/is the design in this area ?
Yours Sincerely
RJ Baars
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
